OpenAI has once again captured attention, this time not due to a new model release. The AI firm, based in San Francisco, is facing allegations of resorting to legal threats to silence critics and influence discussions surrounding California’s recent AI transparency legislation.
The controversy was sparked by Nathan Calvin, the 29-year-old general counsel of a small nonprofit organization named Encode. In a viral post on X, Calvin claimed that OpenAI attempted to weaken California’s Transparency in Frontier Artificial Intelligence Act (SB 53) and employed aggressive tactics to achieve this goal. He asserted that OpenAI even used its legal battle with Elon Musk to portray critics like Encode as being covertly supported by Musk.
Encode, consisting of merely three full-time employees, found itself entangled in a legal dispute after receiving a subpoena in August. The incident, where a sheriff’s deputy arrived at Calvin’s home during dinner, demanding all communications related to OpenAI’s governance and policies, was described as intimidating by Sunny Gandhi, Encode’s vice president of political affairs.
The situation drew reactions from various quarters, including within OpenAI. Joshua Achiam, the company’s head of mission alignment, expressed concerns at potential risks to his career, acknowledging the contentious nature of the issue. Former OpenAI board member Helen Toner criticized the company for its dishonesty and intimidation tactics in policy matters.
In response, OpenAI’s chief strategy officer Jason Kwon defended the company, emphasizing that Encode’s association with Musk raised legitimate questions. Kwon dismissed claims of intimidation and emphasized that subpoenas were standard legal procedures in litigation.
Calvin, however, remained unconvinced, alleging that OpenAI sought to dilute the transparency requirements of SB 53 while it was being drafted. The bill, signed into law by Governor Gavin Newsom in September, mandates AI developers to disclose safety frameworks, report critical incidents, and provide risk assessments to the state.
Amidst the dispute, Encode has refused to comply with OpenAI’s demands, asserting no ties to Musk. Some individuals within OpenAI have called for introspection, urging constructive engagement with critics to uphold the company’s mission. Calvin, though stressed by the ordeal, expressed appreciation for OpenAI’s positive contributions, while questioning the alignment of recent actions with the company’s stated aim of ensuring AI benefits humanity.